[
https://issues.apache.org/jira/browse/MIME4J-259?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15901319#comment-15901319
]
Sebb commented on MIME4J-259:
-----------------------------
As a work-round, the following FromLine setting seems to do the trick:
fromLine("^From \\S+.*\\d{4}\r?(?=\n)")
i.e. explicitly match LF optionally preceeded by CR but don't include the LF in
the match.
The +1 in the code then skips the LF.
> MboxIterator does not handle EOL = CRLF
> ---------------------------------------
>
> Key: MIME4J-259
> URL: https://issues.apache.org/jira/browse/MIME4J-259
> Project: James Mime4j
> Issue Type: Bug
> Reporter: Sebb
>
> The MboxIterator class assumes that the EOL consists of a single character,
> so it increments findEnd by 1 to get the start of the message.
> This causes problems for files which have a CRLF line ending, because the
> generated message contains a leading LF which confuses the parser.
> The iterator should ensure that the entire EOL is skipped.
> However it might be better to keep the From_ line in the returned message.
> The parser can then process it to extract its data.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
